Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    phrases indicating intent or purpose

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "to."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    typ
    -0.69
     folded
    -0.67
    hops
    -0.60
     busted
    -0.60
    need
    -0.58
    hop
    -0.57
     drafts
    -0.55
     rolled
    -0.55
     packs
    -0.55
    Got
    -0.54
    POSITIVE LOGITS
     ensure
    1.18
    ilet
    1.17
    ggles
    1.10
     avoid
    1.08
     satisfy
    1.08
     maximize
    1.08
     minimize
    1.07
     coincide
    1.05
     signify
    1.05
     achieve
    1.04
    Act Density 0.460%

    No Known Activations